• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Strange high vs mid delay measurements for Magnepans

Vasr

Major Contributor
Joined
Jun 27, 2020
Messages
1,409
Likes
1,926
I wonder if anyone has a plausible explanation for this.

I have always used REW to align speakers in a 5.1 setup and it has been great.

I did an experiment with my SMGa mains recently where I measured, with acoustical reference, the tweeter and the mids (it is a 2-way) separately using different sweep ranges (250-550hz for mids and 1000-1300hz for highs). The speakers are placed with tweeters on the outside edge.

On the right speaker the difference between the two ranges with reference to a center speaker acoustical reference is negligible and not very sensitive to toe-in angle within limits.

The left is a whole different story. The mids are coming much faster to the MLP than the highs requiring an extreme toe-in to bring the tweeters forward towards the MLP to balance them in those two ranges. Since they are not bi-amped, I can only provide a delay to the whole signal and so a toe-in is the only way to balance them. But it looks very odd with one speaker with much higher toe-in than the other.

My best guess is that because of the wall geometry behind it, the main reflection of the highs on the outward edge is not reaching the MLP but the secondary reflection is but taking a longer time (order of about 1ms) to reach. The left has a side angled wall starting beyond the speaker (to further left), the right does not.

Unless there is an issue mechanically or electronically with the speaker or the audio chain itself for which I can't think of any reason.

Any thoughts?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,895
Likes
16,714
Location
Monument, CO
Sounds like a wall issue to me. One foot is about 1 ms and is a period of 1 kHz.
 
Last edited:

Jake71

Member
Joined
Aug 17, 2020
Messages
35
Likes
75
I believe some of these have electrolytic capacitors in the crossover, if that's the case I'd check/replace them with film capacitors. Also it's quite common to experience delamination with these speakers. I'd try swapping the two speakers and see what kind of difference it makes, sure it means the tweeters will be closer to the center, but you should still be able to determine if the problem follows one speaker.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,895
Likes
16,714
Location
Monument, CO
Delamination could cause buzzing and such but I do not see how that would cause a phase shift. Ditto cap failure, though if the cap was way, way out of spec, it would affect the crossover frequency and phase.
 

Eurasian

Active Member
Forum Donor
Joined
Nov 2, 2018
Messages
253
Likes
214
Could the elements in the one speaker be wired improperly (polarity of one element) from the factory? If you have access to REW, check for a big suck-out around the crossover point.
 
OP
Vasr

Vasr

Major Contributor
Joined
Jun 27, 2020
Messages
1,409
Likes
1,926
As I had pictured in another thread, I had restored these a few months ago with complete regluing and replacement of the capacitor.

The REW measurements before and after showed a slightly cleaner crossover region and no obvious issues with a complete spectrum scan.

It is a delay issue rather than a FR issue and quite exaggerated at that.

I will switch the speakers and see what happens even though that will bring tweeters inside. Just curious.
 
  • Like
Reactions: wje
OP
Vasr

Vasr

Major Contributor
Joined
Jun 27, 2020
Messages
1,409
Likes
1,926
Measured it after switching speakers.

It is not the speaker, since the earlier left speaker, moved to the right, showed no significant difference between tweeter range and mid range delay measurements and could be corrected to within margin of measurement error with a slight toe-in.

The earlier right speaker which was fine on the right, moved to the left, now showed some delay differences between mid and tweeter driver but not as exaggerated as the previous arrangement with the tweeter on the outside. It could be corrected with a slightly more toe-in arrangement. I suspect that with the tweeter being further away from the side angled wall, the highs weren't as affected as they were earlier with the tweeter closer to the side angled wall on the left side.

Some lessons from this experiment:
1. It is very useful and perhaps a good idea to do separate driver measurements for delay in Magnepans and determine the toe-in to align them for delay. Just pick a range for each driver away from the cross-over region. I used 300hz width sweeps. This should be a better solution than the factory thumb rule.
2. Tweeters outside can be affected by side wall geometry more than the mid drivers. So tweeters in should be used if the room has a geometry where this becomes an issue. Note that it does affect the sound-stage and some may already have them with the tweeters on the inside as their preference.
 
Top Bottom